한국전자통신학회논문지 (The Journal of the Korea institute of electronic communication sciences)

한국전자통신학회 (Korea Institute of Electronic Communication Science)
권호 표지
DOI QR코드

DOI QR Code

서포트벡터머신 기반 PVDF 센서의 결함 예측 기법

Fault Detection Technique for PVDF Sensor Based on Support Vector Machine

  • 김승욱 (한양대학교 융합국방학과) ;
  • 이상민 (고려대학교 기계공학부)
  • 투고 : 2023.08.14
  • 심사 : 2023.10.17
  • 발행 : 2023.10.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구에서는 구조물 건전성 모니터링에 널리 활용되고 있는 PVDF(: Polyvinylidene fluoride) 센서에 나타날 수 있는 결함을 실시간으로 분류 및 예측하기 위한 방법론을 제안하였다. 센서 부착 환경에 따라 나타나는 센서의 결함 유형을 분류하였고, 임팩트 해머를 이용한 충격 시험을 수행하여 결함 유형에 따른 출력 신호를 획득하였다. 결함 유형에 따른 출력 신호간의 차이를 식별하기 위해 이들의 시간영역 통계 특징을 추출하여 데이터 집합을 구축하였다. 머신러닝 기반 분류 알고리즘들 중 센서 결함 유형 감지에 가장 적합한 알고리즘 선정을 위해 구축한 데이터 집합의 학습 및 이에 따른 결과를 분석하였고, 이들 중 SVM(: Support vector machine)이 가장 높은 성능을 보임을 확인하였다. 선정된 SVM 알고리즘의 추가적인 정확도 향상을 위해 하이퍼 파라미터 최적화 작업을 수행하였으며, 결과적으로 92.5%의 정확도로 센서 결함 유형을 분류하였고 이는 타 분류 알고리즘에 비하여 최대 13.95% 높은 정확도를 보였다. 본 연구에서 제안한 센서 결함 예측 기법은 PVDF 센서뿐만 아니라 실시간 구조물 건전성 모니터링을 위한 다양한 센서의 신뢰성을 확보하기 위한 기반 기술로 활용될 수 있을 것으로 사료된다.

In this study, a methodology for real-time classification and prediction of defects that may appear in PVDF(Polyvinylidene fluoride) sensors, which are widely used for structural integrity monitoring, is proposed. The types of sensor defects appearing according to the sensor attachment environment were classified, and an impact test using an impact hammer was performed to obtain an output signal according to the defect type. In order to cleary identify the difference between the output signal according to the defect types, the time domain statistical features were extracted and a data set was constructed. Among the machine learning based classification algorithms, the learning of the acquired data set and the result were analyzed to select the most suitable algorithm for detecting sensor defect types, and among them, it was confirmed that the highest optimization was performed to show SVM(Support Vector Machine). As a result, sensor defect types were classified with an accuracy of 92.5%, which was up to 13.95% higher than other classification algorithms. It is believed that the sensor defect prediction technique proposed in this study can be used as a base technology to secure the reliability of not only PVDF sensors but also various sensors for real time structural health monitoring.

키워드

참고문헌

  1. Y.-H. Huh, J.I. Kim, J.H. Lee, S.G. Hong and J.H. Park, "Application of PVDF Film Sensor to Detect Early Damage in Wind Turbine Blade Components," Procedia Eng., vol. 10, 2011, pp. 3304-3309. https://doi.org/10.1016/j.proeng.2011.04.545
  2. D.-W. Keum and J.-D. Kim, "Measurement of Apnea Using a Polyvinylidene Fluoride Sensor Inserted in the Pillow," Journal of Sensor Science and Technology, vol. 27, no. 6, 2018, pp. 8682-8690.
  3. D. Tibaduiza, M. Anaya, E. Forero, R. Castro and F. Pozo, "A Sensor Fault Detection Methodology applied to Piezoelecrtic Active Systems in Structural Health Monitoring Application," IOP Conf. Ser.: Mater. Sci. Eng., vol. 138, no. 1, 2016, p. 012016. https://doi.org/10.1088/1757-899X/138/1/012016
  4. G. Park, C. R. Farrar, F. L. di Scalea, and S. Coccia, "Performance Assessment and Validation of Piezoelectric Active-Sensors in Structural Health Monitoring," Smart Mater. Struct., vol. 15, no. 6. 2006, pp. 1673-1683. https://doi.org/10.1088/0964-1726/15/6/020
  5. T.-C. Huynh, T.-D. Nguyen, D.-D. Ho, N.-L. Dang, and J.-T. Kim, "Sensor fault diagnosis for impedance monitoring using a piezoelectric-based smart interface technique," Sensor, vol. 20, no. 2, 2020, pp. 510-530. https://doi.org/10.3390/s20020510
  6. B. Samanta, "Gear fault detection using artificial neural networks and support vector machines with genetic algorithms," Mech. Syst. Signal Process., vol. 18, no. 3, 2004, pp. 625-644. https://doi.org/10.1016/S0888-3270(03)00020-7
  7. T. W. Rauber, F. de A. Boldt, and F. M. Varejao, "Heterogeneous feature models and feature selection applied to bearing fault diagnosis," IEEE Trans. Ind. Electron., vol. 62, no. 1, 2015, pp. 637-646. https://doi.org/10.1109/TIE.2014.2327589
  8. N. S. Altman, "An Introduction to Kernel and Nearest-neighbor Nonparametric Regression," The American Statistician, vol. 46, no. 3, 1992, pp. 175-185. https://doi.org/10.1080/00031305.1992.10475879
  9. L. Rokach and O. Maimon, "Top-down Induction of Decision Trees Classifiers-a Survey," IEEE Transactions on Sustems, Man and Cybernetics Part C : Applications and Reviews, vol. 35, no. 4, 2005, pp. 476-487. https://doi.org/10.1109/TSMCC.2004.843247
  10. S.-J. Yu, "A Study on Recommendation Method for Real Estate Using Naive Bayes Classification," The Journal of Korean Institute of Information Technology, vol. 17, no. 10, 2019, pp. 115-120. https://doi.org/10.14801/jkiit.2019.17.10.115
  11. S. U. Jan, Y.-D. Lee, J. Shin, and I. Koo, "Sensor fault classification based on support vector machine and statistical time-domain features," IEEE Access, vol. 5, 2017, pp. 8682-8690. https://doi.org/10.1109/ACCESS.2017.2705644
  12. D. Ai, H. Luo, and H. Zhu, "Diagnosis and validation of damaged piezoelectric sensor in electromechanical impedance technique," Journal of Intelligent Material Systems and Structures, vol. 28, no. 7, 2017, pp. 837-850. https://doi.org/10.1177/1045389X16657427
  13. A. M. AY, Y. Wang and S. Khoo, "Signal processing for time domain analysis of impact hammer test data," 8th European Workshop on Structural Health Monitoring, EWSHM, vol. 1, 2016, pp. 628-637.
  14. X. Deng and V. Giurgiutiu, "Impact Monitoring and Fault Detection Using Piezoelecrtic Transducers and Wavelet Analysis," 53rd Meeting of the Society for Machinery Failure Prevention Technology, Virginia Beach, VA., 1999 pp. 167-172.
  15. H. Peng, F. Long, and C. Ding, "Feature Selection Based on Mutual Information: Criteria of Max-Dependency, Max-Relevance, and Min-Redundancy," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 8, 2005, pp. 1226-1238. https://doi.org/10.1109/TPAMI.2005.159
  16. S. Arlot and A. Celisse, "A survey of cross-validation precedures for model selection," Stat. Surv., vol. 4, 2010, pp. 40-79. https://doi.org/10.1214/09-SS054
  17. B. Shahriari, K. Swersky, Z. Wang, R. P. Adams and N. de Freitas, "Taking the Human out of the Loop: a Review of Bayesian Optimization," Chinese Journal of Evidence-Based Medicine, vol. 14, no. 10, 2014, pp. 1270-1275.
  18. T. Fawcett, "An Introduction to ROC Analysis," Pattern Recognit. Lett., vol. 27, no. 8, 2006, pp. 861-874. https://doi.org/10.1016/j.patrec.2005.10.010